Top
Best
New

Posted by Liwink 1 day ago

DeepSeek v4.1 Flash(twitter.com)
https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash
976 points | 557 commentspage 7
lwansbrough 1 day ago|
Significant jump in pricing. V4 Flash was $0.16/M out, 4.1 is $1.20/M.
svantana 1 day ago||
I think you're comparing to third party prices, deepseek's prices hasn't changed with this release. Also, $1.2 is the peaktime price.

https://api-docs.deepseek.com/quick_start/pricing/

trq01758 1 day ago|||
Never saw $0.16 for 1M output tokens - it was $0.28 a month ago, $0.66 off-peak and $1.32 peak last week, now it is reduced a bit to $0.6 and $1.2
lwansbrough 1 day ago||
Was looking at OpenRouter, I guess it’s wrong.
dakolli 1 day ago||
incorrect, no idea where you're getting this pricing. Also, output does not matter. its 10% of the cost.
esafak 18 hours ago||
It is fast! https://artificialanalysis.ai/models/deepseek-v4-1-flash

https://deepseek.com/en/news/deepseek-v4-1-flash/

peter_d_sherman 22 hours ago||
>"Smaller KV cache. Bigger savings.

Compared with the previous generation, V4.1-Flash’s KV cache needs just:

o 1/4 the HBM

o 1/8 the SSD storage

Cache-hit charges often account for a large share of agent costs. Compressing the cache cuts those costs significantly."

It makes one wonder as to just how far an LLM's KV cache could theoretically be shrunk before losing significant functionality...

bellowsgulch 23 hours ago||
OpenCode Go referral code, if you want to try it. https://opencode.ai/go?ref=QDJQMTGP5Q
Translationaut 21 hours ago|
Still 10$/month with this referral code?!
bellowsgulch 19 hours ago||
Yeah, unfortunately. But you get an additional $5.
gigatexal 1 day ago||
I’m all in on Chinese models, Deepseek especially given how cheap it is. It’s also really solid and comparable in real world use to a sonnet for my work.
jonplackett 1 day ago||
Can we just never link to X posts as the main link.
small_model 22 hours ago||
No, that is called censorship
igravious 22 hours ago|||
https://news.ycombinator.com/from?site=twitter.com

There have been 34 Twitter/X link submissions in the past day, ~that's 12,000 submissions a year.

If your reason is that you have to be logged in to use it properly then I'd nearly agree with you. If it's for any other reason, how about no?

jhonof 21 hours ago||
The login issue is extremely annoying, at least mandating an xcancel link would fix that.
nunodonato 1 day ago||
yes, please. Especially now that xcancel is gone
addandsubtract 1 day ago||
Xcancel is back: https://news.ycombinator.com/item?id=49588988
arjie 1 day ago||
What in the world. A point release with 2x the parameters and a different architecture? Jesus. Can’t run this kind of thing on 2x RTX Pro 6k at decent speed. I need to reconfigure my hardware. Massive disappointment on that front. Bloody hell. Glad I didn’t get a DGX Station.

No wonder they retired the Pro model in favour of this.

codedump 1 day ago||
[dead]
kryzz-ai-bo 23 hours ago|
[dead]
More comments...